Download Cooling Down Economy with Fiscal Policy in a Monetary Union

Cooling Down Economy with Fiscal Policy...
245
Ph.D. Paweł Młodkowski
Visiting Professor
Department of Economics
SOKA University
Cooling Down Economy with Fiscal Policy
in a Monetary Union
INTRODUCTION
Economic interaction is very often perceived as a process of a sequential
giving up autonomy in economic policy of national authorities and delegating it
to supranational level. Under this approach, the monetary union is the final stage
of economic integration, when member states move monetary policy decisions
to a common institution (a common central bank). Despite many potential
benefits flowing from using a common currency, it was already pointed out by
the classical OCA theory that there is a relationship between macroeconomic
losses and several criteria that characterize member economies. These were
connected with the probability and severity of asymmetric shocks and any
responses by economic policies.
When the stage of full monetary integration is achieved, the only
autonomous policy at the national level is the fiscal policy. This means that it
becomes relatively more important for achieving national economic policy
goals, for example when cooling down is required to converge business cycles
in a monetary union. Despite a very good elaboration of the role and
consequences of monetary policy actions in the DSGE-class models, there is
neither similar consensus nor analytical framework in case of the fiscal policy.
The European Central Bank and other central banks outside Europe use large
macromodels that allow for forecasting consequences of their decisions
[Wouters & Smets 2004, Staub & Tchakarov 2007]. They are, however, not able
to capture consequences of the independent fiscal policies. These models are
based on extensive microfoundations covering a representative agent with the
infinite planning horizon. They assume, however, that the fiscal policy does not
influence demand or internal equilibrium. These two assumptions can be found
invalid on the empirical ground [Kumhof & Laxton 2007a]. Therefore there is
a need for developing modelling frameworks in which dynamic effects of fiscal
policy would be properly captured and used for forecasting consequences of
246
PAWEŁ MŁODKOWSKI
a policy mix. This is true not only in the case of a monetary union, but for any
country case in which independent economic policies are implemented.
However, the problem of the role of the fiscal policy in converging business
cycle phases among currency union member states is of superior importance in
monetary union economics.
The aim of this paper is to present a modified version of the GIMF (Global
Integrated Monetary and Fiscal model) created by the International Monetary
Fund. The modification takes form of adjustments reflecting some additional
assumptions and mechanisms present under full monetary integration (lack of
monetary policy instruments in cooling down domestic economy). The
empirical studies that were utilizing the original GIMF model recognized so far
the following features [Leight 2008]:
− how stronger response of the fiscal policy in consecutive business cycle
phases can decrease the scope of necessary monetary policy response,
− to what extent stronger fiscal policy response in consecutive business cycle
phases induces substitution effect: inflation to output,
− in what way efficiency of fiscal policy in increasing macroeconomic
stability depends on interaction with monetary policy and lags in its
implementation.
This paper presents the methodology, discusses theoretical background and
formulates suggestions in regard to implementing the modified version of the
GIMF model and further research directions.
MODIFICATION OF THE GIMF FRAMEWORK
FOR A CURRENCY UNION SETTING
The GIMF model belongs to a modern class of open economy models. This
is a model of general equilibrium, which allows for analyzing both monetary
and fiscal policies (policy mix). In particular, the assumptions result in not
conforming to the Ricardian equivalency. The Ricardian equivalency is not
present due to the characteristics of the representative agents. The overlapping
generations have a finite time horizon. As a consequence, the fiscal expansion
affects permanently their wealth and they do not expect the necessity of bearing
costs in a form of higher taxes in the future. In addition, productivity of labor
that decreases with age means that workers discount future increases in income
taxes. The older they are the probability of being affected by increased taxes is
lower. Consumers do not have access to financial markets, which prevents
consumption smoothing. It results in a direct and a very strong impact of any
Cooling Down Economy with Fiscal Policy...
247
changes in income taxes on demand. This condition / assumption must be
perceived as the most severe drawback of the presented model. Further
theoretical research and developing models of this class require departing from
this assumption since it is inconsistent with the reality.
In this model there are two taxes: the income tax and the tax on investment.
They are causing distortions because both consumption and investment depend
on relative prices. This assumption results in the fact that they are directly
affected by any changes in the tax rates.
The original version of the GIMF model assumes that public spending is
productive. In particular, this refers to spending on infrastructure and increasing
productivity of production factors in the private sector this way. Real rigidity
refers to customary preferences in the area of consumption. This real rigidity
induces inertia of investment and generates costs of any changes in imports.
Nominal rigidities are embodied by sticky Phillips curves in each sector
[Kumhof & Laxton 2007a].
The government is able to set the ratio of the budget BALance to the GDP
and to raise additional tax revenue. The fiscal policy rule is defined as follows:
BALt/GDPt = α + β (Tt – Tt*/GDPt)
(1),
where BALt/GDPt is the ratio of the budget BALance to the GDP,
β – is a parameter describing deficit response (fiscal policy response) to
changes in the public revenue from the income taxes.
When β=0, then BALt/GDPt is maintained always at the level „α”. Tt is
the actual tax revenue while T* is the tax revenue in equilibrium.
Procyclality of this rule is present both at β = 0 and β < 0. With the increase
of β above zero, more and more additional income tax revenue is saved.
When β = 1, the response of the fiscal policy can be recognized as a
structural deficit, since 1% increase of the tax revenue (as the % of the GDP)
leads to 1% increase in the BALance. Conducting fiscal policy is simply
based on adjusting public spending (G) or the tax revenue (Tt). For β > 1 the
fiscal policy becomes counter-cyclical and tax smoothing is present.
Deficits (negative BALances) appear in recessions and accumulated public
debt is repaid during booms.
The monetary policy, according to present solutions, when based on
inflation targeting, is conducted in a form of adjustments of nominal interest
rates. Modelling of this element is very often based solely on expectations and
assumption of some inertia [Lopez 2003]. Adjusting the formula for monetary
policy to institutional framework at the European Central Bank (ECB) yields the
following equation:
248
PAWEŁ MŁODKOWSKI
µi
(
it = it −1 × r × Π t + 4
*
t
)
1− µi t
 Π t +4 
×  Et * 
 Π t +4 
(1− µi )µΠ
(2),
where it is the nominal interest rate set by the central bank,
Π* – is the inflation target defined as a joint inflation for the forecast period
of 4 quarters (Πt+4 = Πt+1 · Πt+2 · Πt+3 · Πt+4),
E – reflects expectations formulated on the basis of information available at
time „t”.
The coefficient µi є [0 ; 1) reflects nominal interest rate inertia.
If µi=0, equation (2) implies that when the inflation forecast exceeds the
target by 1 percentage point, the nominal interest rate increases by 1+ µπ .
The equilibrium real interest rate (rt*) is endogenous, and is determined
by the global market for loanable funds, as well as a country-specific risk
premium.
The real interest rate at equilibrium is a dependent variable, set at the global
loan market and includes specific risk premium for a monetary union. In case of
the Eurozone this risk premium is relatively small, but its theoretical
relationship with the fiscal policy in the model requires this element to be
introduced. This premium is a difference between the monetary union interest
rate and the rest of the world interest rate subject to expected changes of the
exchange rate.
it = itRW × Etε t +1 (1 + ϕt )
itRW–
(3)
where
is the nominal interest rate in the rest of the world,
εt+1 – is the future rate of change of the nominal exchange rate, and
φt is the monetary union-specific risk premium.
In an open economy setting the interest rate must comply with the
uncovered interest rate parity. The central bank ex ante sets its interest rate at a
certain level, but the market forces (via uncovered interest rate parity) induce
appropriate capital flows and any difference between the ex ante interest rate set
by the central bank and the global interest rate (plus risk-specific premium) and
ex post interest rate disappears. The aim is to introduce the impact of fiscal
policy in the model by assuming that the risk premium depends on fiscal policy.
Therefore (ex post) the central bank will adjust the interest rate or allow for
appreciation/depreciation of the common currency – as a result of fiscal policy
actions.
This element must be therefore associated with the fiscal policy to allow
modelling of agents’ response to the difference in rates of return on domestic
Cooling Down Economy with Fiscal Policy...
249
(i.e. union) and foreign assets. For the framework consistent with nominal
convergence criteria in the EMU the following formula is required:
ϕ t = δ1 +
δ2

PDt
 0.6 −
GDP
t

(4).
δ3



The first element in brackets (0.6 => 60%) reflects the maximum level of
public debt that is allowed in monetary union member states. The bigger the
difference between the limit and the actual level of public debt, the higher the
flexibility of the fiscal policy [Młodkowski 2007]. In addition, it results in
decreasing the specific risk premium and drives the union-wide interest rate
closer to the global one. Parameters δ1, δ2, δ3 reflect the specific features of
the monetary union economy. δ1 is the premium for a specific risk
independent from budgetary issues. δ2 is always positive and reflects reaction
observed in empirical data between risk premium and the level of the public
debt [Arora & Cerisola 2001] – positive impact of debt on risk premium in the
nominal interest rate. δ3 is also positive and it sets the scope of convexity of the
function φt.
For the households (in the monetary union and in the rest of the world) 15year planning horizon is assumed and decreasing productivity at 5% yearly
[Kumhof & Laxton 2007b]. Coefficients describing fiscal policy and impact of
public investment on the GDP must be estimated on the basis of separate models
or be taken from reviews of studies utilizing meta-analysis as in Ligthart &
Suarez [2005].
The model presented here allows for testing different levels of the „β”
parameter that reflects the nature of the fiscal policy reaction in each of the
business cycle phases. Available tools of influencing business activity cover
income tax, consumption tax (VAT) and tax on investing in financial
instruments.
In the steady state the fiscal balance is supposed to be maintaining the ratio
PDt/GDPt at the stable level. This means proportionate fluctuations of debt and
output according to the rate of change of the nominal GDP:
∆GDP  PD 
 DEF 
×

=

 GDP  1 + ∆GDP  GDP 
(5).
In a hypothetical case for the EMU, if the PDt/GDP ratio in equilibrium is
for the EMU countries at 40% (on average) and the average nominal GDP
growth rate is around 4,23% (table 2), then the deficit in the Eurozone must not
exceed 1,62% of the GDP (on average). Adjusting the parameter that is
250
PAWEŁ MŁODKOWSKI
responsible for nominal interest rate inertia (µi) one can refer to paper by
Clarida, Gali, Gertler [1998], who estimated this parameter also for the German
economy at 0,8 (monthly data) and 0,51 (quarterly data).
On the basis of results obtained by Rowland and Torres [2004] one can
assume that in the EMU growth of PDt/GDPt by one percentage point would
lead to change in specific risk premium by 5–7 basis points.
On the foundations of the model for the Eurozone it becomes possible to
study fiscal policy effects when cooling down economic activity is conducted as
a response to a positive asymmetric shock. At the beginning it is necessary to
point out specific conditions prevailing in the EMU for fiscal policies. Because
of high levels of public debt in the biggest Eurozone economies the flexibility of
fiscal policy is low. This element is, however, of minor importance when
studying a response to a positive economic shock.
Cooling down process is defined as decreasing the deficit (DEFt) and debt
(PDt) that leads to increasing potential scope of fiscal policy response in the
future to negative asymmetric shocks. Comparing the assumptions and
parameters describing growth rates of the nominal GDP and the limits for debt
and deficit with the current situation of the EMU member states one can
conclude that from the very beginning of the Eurozone fiscal policy was
expansionary. This resulted in the unstable PD/GDP ratio and executing
influence on premium on specific risk for all EMU issuers. This observation is
supported by information on interest rates in table 1.
Table 1. Interest rates (after removing inflation premium)
and the average fiscal balance in the EMU
EMU interest rates
EMU fiscal
position
1999
2000
2001
2002
2003
2004
2005
2006
3.43
3.75
4.08
3.83
2.10
3.15
2.14
3.88
-1.7
-0.4
-1.7
-2.1
-2.4
-2.5
-2.2
-1.4
2007
-1.1
Source: International Financial Statistics, International Monetary Fund, June 2008 (for interest
rate) and the European Central Bank database (for fiscal position).
As an important impediment in stabilizing PD/GDP ratio in the Eurozone
one can recognize low nominal GDP growth rates (table 2).
Table 2. Euro 12 – Growth rate of the gross domestic product at market prices 1999–2007
1999
2000
2001
2002
2003
2004
2005
2006
2007
4.58%
5.21%
4.35%
3.50%
2.94%
4.04%
3.70%
4.79%
4.92%
Source: European Central Bank database.
Cooling Down Economy with Fiscal Policy...
251
The only available solution is a cut in public spending to reach the deficit
below 1.62% GDP (this is observed in 2006&2007). Only after arriving at such
stance, the debt/GDP ratio will become stable and will start to drop (as indicated
in figure 1 for 2006&2007).
Figure 1. The average debt-to-GDP ratio for the 15 EuroArea
members 2000–2008 (quarterly)
Source: European Central Bank database.
However, it is already the process of achieving this level of deficit can be
perceived as restrictive fiscal policy. The proposed model may help in
answering questions about any potential consequences.
Figure 2. Estimated relationship of φt (risk premium) and PD/GDP
for EURO AREA from 2000 to 2008
Source: Author.
252
PAWEŁ MŁODKOWSKI
As expected, cuts in public spending lead in the GIMF model to an expedite
decrease in demand. In the long run theoretically there is a relationship of GDP
returning to the initial level. This is because of savings on interest payments and
possibility of decreasing taxes, when PD/GDP drops.
The mentioned decrease in demand in a monetary union member state has
its impact on prices, which were growing faster in a country hit by a positive
demand shock. The common monetary policy will not respond to this antiinflation impulse generated by a restrictive fiscal policy. There is actually no
way it could be done due to the impossible trinity. Instead, one should expect
increase in the effectiveness of the common monetary policy resulting from
converging business cycle phases in member economies. Decreasing domestic
absorption leads to improvement in trade balance and the current account. If one
assumes that the internal equilibrium was associated with zero net exports, then
an additional effect of the restrictive fiscal policy would be accumulation of
foreign exchange reserves and increase of domestic savings. This is a result of a
commonly accepted relationship based on an equivalency derived from national
accounts – twin deficits. In addition, one can notice the fact that this is a feature
that distinguishes the GIMF-class models from the standard open economy
models based on Ricardian assumptions.
CONCLUSION
The proposed model is a simple modification of a standard GIMF created
by the IMF. A very important feature is the lack of monetary policy response to
fiscal contraction. It results in omitting some mechanisms included in the
standard GIMF version [Leight 2008]. However, even without the support in the
form of decreasing nominal interest rates (Figure 2), there are some positive
effects of decreasing public spending. In the medium and the long run these
benefits cover increase in savings, drop in interest payments and potential for
decreasing taxes on income and capital. Decreasing taxes (and the specific risk
premium for union-issuers) results in increasing labour supply, investment, the
GDP and consumption. The future directions for studies and developing GIMF
models for a monetary union should focus on calibrating parameters and
implementing all nominal convergence criteria.
BIBLIOGRAPHY
Arora V., Cerisola M. (2001), How Does U.S. Monetary Policy Influence Sovereign
Spreads in Emerging Markets?, IMF Staff Papers Vol. 48, No 3, pp. 474 – 498.
Clarida R., Gali J., Gertler M. (1998), Monetary Policy Rules in Practice: Some
Cooling Down Economy with Fiscal Policy...
253
International Evidence, „European Economic Review”, Vol. 42, pp. 1033–67.
Kumhof M., Laxton D. (2007), A Party Without a Hangover? On the Effects of U.S.
Government Deficits, IMF Working Paper Nr 07/202, International Monetary Fund,
Washington D.C.
Leight D. (2008), Achieving a Soft Landing: The Role of Fiscal Policy, IMF Working
Paper Nr. 08/68, IMF, Washington D.C.
Ligthart J., R. M. Suárez (2005), The Productivity of Public Capital: A Meta Analysis,
manuscript, University of Groningen, and Tilburg University.
López, M. (2003), Efficient Policy Rules for Inflation Targeting in Colombia,
„Borradores Semanales de Economía”, No. 240.
Młodkowski P. (2007), Fiscal Policy Flexibility in a Monetary Union, Folia Economica
Cracoviensia, Vol. XLVIII 2007, pp. 25 – 46.
Młodkowski P. (2008), Microfoundations in Monetary Union Modeling, International
Advances in Economic Research, Vol. 14, No. 4, November 2008, pp. 481–482.
Summary
The paper offers a modification to a simple theoretical framework of the DSGE class model
for an open economy with extended microfoundations. It can be used for an analysis of an
economy of a monetary union when a member state is hit by an asymmetric shock. The focus is on
the response of national fiscal policy and its efficiency under nominal convergence criteria that
apply for central government deficit and public debt. Due to non-Ricardian assumptions regarding
representative agents it becomes possible to capture reaction of consumption and investment to
alterations in the level of taxation. The proposed modification of the standard GIMF framework is
quite simple and takes a form of introducing a limit for public debt (defined as % of GDP) and
removing direct interaction with monetary policy that would otherwise (without full monetary
integration) decrease social cost of fiscal contraction.
Schładzanie gospodarki poprzez politykę fiskalną w Unii Walutowej
Streszczenie
W opracowaniu proponuje się modyfikację prostych teoretycznych ram modeli klasy DSGE dla
gospodarki otwartej o rozszerzonych podstawach mikroekonomicznych. Mogą one znaleźć zastosowanie
w analizach gospodarki Unii Walutowej w sytuacji, kiedy państwo członkowskie znalazło się w sytuacji
asymetrycznego szoku. Uwaga ogniskuje się na możliwych działaniach w ramach narodowej polityki
fiskalnej i ich efektywności w sytuacji ograniczeń w postaci nominalnych kryteriów konwergencyjnych
odnoszących się do deficytu budżetu państwa i długu publicznego. W oparciu o pozaricardiańskie
założenia dotyczące reprezentatywnych agentów możliwe staje się uchwycenie reakcji konsumpcji
i inwestycji na zmiany w poziomie opodatkowania. Proponowana modyfikacja ram standardów GIMF
jest nieskomplikowana i przyjmuje kształt wprowadzenia ograniczeń dla zadłużenia publicznego
(określonego jako procent PKB) oraz usunięcia bezpośrednich związków z polityką monetarną, która
w przeciwnym wypadku (bez pełnej integracji monetarnej) zmniejszyłaby społeczne koszty ograniczeń
fiskalnych.